-
Notifications
You must be signed in to change notification settings - Fork 14
CLI improvements #61
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
CLI improvements #61
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Pull request overview
This pull request enhances the CLI to support continuing from a checkpoint with a different name than the current training run, and removes epoch-related data from logging in preparation for data mixtures support.
Changes:
- Modified
--continueCLI option to accept string arguments (checkpoint paths or step numbers) instead of just integers - Removed
epochparameter from logging functions and replaced epoch-based progress calculation with step-based progress - Added
set_final_stepmethod to enable accurate progress tracking
Reviewed changes
Copilot reviewed 3 out of 3 changed files in this pull request and generated 6 comments.
| File | Description |
|---|---|
| train.cpp | Updated checkpoint continuation logic to support paths and directory names; removed epoch parameters from function calls; added final step tracking |
| src/training/logging.h | Updated function signatures to remove epoch parameters; renamed mRemainingTokens to mFinalStep; added set_final_step method |
| src/training/logging.cpp | Implemented step-based progress calculation; removed epoch from log output; added set_final_step implementation |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Pull request overview
Copilot reviewed 4 out of 4 changed files in this pull request and generated 2 comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Add the ability to continue from a checkpoint with a different name than the current training run.
Remove "epoch" data from logging, it isn't really useful and won't generalize well once we add data mixtures.